Summary: Use of tumor vaccines based on dendritic cells has generated tremendous interest in the field of cancer immunotherapy. Dendritic cells (DC) are professional antigen presenting cells which are relatively rare in normal tissues, but have the potential to enhance tumor-specific T cell responses upon expansion. One approach currently in clinical trials involves fusion of dendritic cells with autologous patient tumor cells to generate tumor cell-dendritic cells. Both an autologous approach and and allogeneic approach are proposed for clinical trials using tumor-cell DC hybrid vaccines. In the autologous approach, patient DC are expanded from peripheral blood monocytes using GM-CSF and IL-4, then fused with tumor cells and reinjected. In the allogeneic approach, DC are prepared from blood bank materials and fused with tumor cells prior to injection. These two approaches are conceptually different in the way that CD4+ T cells provide T cell help to anti-tumor CD8+ T cells. In the autologous approach, CD4+ cells will recognize tumor antigens presented in the context of self MHC and provide T cell help to anti-tumor CD8+ T cells. In the allogeneic approach, T cell help is provided by CD4+ cells recognizing allogeneic MHC class II (allospecific CD4+ cells). To examine the requirements for T cell help we will employ a murine tumor model system. E.G7 cells are a derivative of the mouse lymphoma EL4 (H-2b) which has been transfected with ovalbumin (OVA). OVA is an advantageous model tumor antigen because both MHC class I and class II epitopes have been defined. OT-I mice are transgenic for the CD8+ T cell receptor recognizing OVA in the context of H-2 Kb. However, if wild-type B6 mice are injected with E.G7 cells, and then T cells from OT-I mice are adoptively transferred at appropriate doses, the OT-I cells are not primed by tumor, and the mice will succumb to tumor burden. CD4+ T cell help is required for anti-tumor priming.To efficiently prime mice for tumor rejection, we propose using DC-tumor cell hybrids analogous to the human clinical trials. To model autologous tumor-specific CD4+ T cell help, we will prime mice with hybrids between E.G7 cells and syngeneic B6 bone-marrow derived DC (BM DC). To generate allospecific CD4+ T cell help, we will make hybrids between E.G7 cells and BM DC from B10.MBR mice. This strain expresses Kb, so DC can present the MHC class I restricted OVA tumor antigen to CD8+ T cells from OT-I mice. However, in B10.MBR mice, the MHC class II (I-A and I-E) are H-2k haplotype, so DC will elicit allospecific CD4+ T cells rather than CD4+ cells specific for OVA in the context of H-2b MHC class II. These two murine systems will be ideal to compare CD4+ T cell help provided by allospecific T cells with help provided by autologous tumor-specific T cells. In the last year, we have only procurred mice reagents for the project but have not initiated studies.